FISHERY BULLETIN: VOL. 69. NO. 4 



the maximum target strength slightly greater 

 than at 30 kHz. Although these measurements 

 have produced some useful results on the gen- 

 eral changes of target strength with fish size 

 and frequency for aspects of special interest, 

 little has been learned about the fine-scale 

 changes or about the differences among species. 



The use of a third technique jiermits investi- 

 gation into the nature of the echo-formation 

 process either by dissection or modeling. By dis- 

 section, researchei's have discovered which parts 

 of a fish are the major acoustic reflectors. By re- 

 moving the swim bladders from a number of 

 perch, Jones and Pearce (1958) determined that 

 the gas-filled swim bladder accounts for approx- 

 imately .'50 ''r of the dorsal and side-aspect target 

 strengths of perch at L \ = 4. Hence, the swim 

 bladder is an imi)ortant contributor to target 

 strength for L/\ values in the interference re- 

 gion. By systematically removing various parts 

 of a skipjack tuna, Volberg (1963) found that 

 appreciable echoes could be obtained from either 

 the skeleton or a piece of flesh. Diercks and 

 Goldsberry (1970) have indicated the possibility 

 that scales may also be an imiiortant contributor 

 to the target strength of a fish at certain fre- 

 quencies. Unfortunately, they did not remove 

 any scales and their hypothesis is based on con- 

 siderations of the directivity of the scales as an 

 array of scatterers. 



An adjunct to the determination of the parts 

 of the fish which are acoustically imjjortant is 

 the determination of the acoustic impedance or 

 reflection coeflicient of these parts. The reflec- 

 tion coeflicient is defined as it was previously for 

 two semi-infinite media. The impedance of the 

 gas in the swim bladder is readily determined, 

 and the reflection coefficient for the swim bladder 

 is approximately — 1. Determination of the 

 acoustic impedance of fish bone or flesh is dif- 

 ficult and care must be taken to insure valid 

 measurements. Shishkova (1958) measured the 

 density of and speed of sound in flesh from a few 

 species of fish and determined the reflection co- 

 eflicient in fresh water to be about 0.05. Haslett 

 (1962b) used a different technique to indirectly 

 measure the reflection coeflicieiits of flesh and 

 bone from haddock and cod. He found the re- 

 flection coeflicient of flesh in fresh watei- to be 



about 0.05, in seawater to be about 0.02, and the 

 reflection coeflicient of bone to be about 0.25. 



Using these values for the reflection coefll- 

 cients and his "standard fish dimensions," Has- 

 lett (1962c, 1964) has modeled fish bodies, back- 

 bones, and swim bladders. Utilizing rubber 

 ellipsoids to model the fleshy body of the fish, 

 he found that the number of lobes obtained in 

 polar plots for the ellipsoids and for actual fish 

 agreed fairly well, that is, with less than 509f 

 error, but that the target strengths obtained for 

 the models were considerably lower than those 

 obtained for the fish. Using rubber and plastic 

 cylinders to model the backbone and copper cyl- 

 inders to model the swim bladder, Haslett has 

 examined variations in the target strength of 

 these models as frequency, size, and aspect are 

 varied. A brief summary of Haslett's work for 

 side aspect is shown in Figure 3. Along with 

 his data for the acoustic cross-sections of stickle- 

 backs and guppies, apjM'oximations to the 

 acoustic cross-sections of the swim bladder, body, 

 and backbone are given. The various curves for 

 each component were determined by Haslett 

 (1965) using his reflection coeflicients and the 

 results of his modeling experiments and depend 



10 



FiGl'Rlv 'A. — Sido-aspect acoustic cross-sections dptprmincd 

 by Haslett. 



708 



